feat: implement 0x40 - BlockHash Opcode

[Eikix]

Since	Group
Frontier	Block Information

Index 1 is top of the stack. See PUSH.

Stack input

1. blockNumber: block number to get the hash from. Valid range is the last 256 blocks (not including the current one). Current block number can be queried with NUMBER.

Stack output

1. hash: hash of the chosen block, or 0 if the block number is not in the valid range.

Examples

	Input	Output
1	599423545	0x29045A592007D0C246EF02C2223570DA9522D0CF0F73282C79A1BC8F0BB2C238

Error cases

The state changes done by the current context are reverted in those cases:

• Not enough gas.
• Not enough values on the stack.

[Eikix]

Important update: we are gathering some bugs in the Kakarot v0 codebase, we need to make sure each issue and each PR in Kakarot-ssj is aware of the lists of known bugs. Look at this link everytime you take an issue and check your issue isn't targeted by a known bug.

[Eikix]

Important update: we are gathering some bugs in the Kakarot v0 codebase, we need to make sure each issue and each PR in Kakarot-ssj is aware of the lists of known bugs. Look at this tracking issue everytime you take an issue and check your issue isn't targeted by a known bug. Will add this reminder in many places to make sure we keep track of known bugs.

[Eikix]

As per
https://docs.starknet.io/documentation/architecture_and_concepts/Smart_Contracts/system-calls-cairo1/#get_block_hash

What should we do for this ~10 block lag?

My proposal:
Start by implementing the function using the syscall, as it's the most straightforward way.
Then for the N-10 to N problem, we can manage using error handling and some hook calling to the sequencer (madara?) -> longer term